Can I set a Luminance filter offset for focus?


I have read the thread about the problems associated with long focal length focusing and using an RC at 1600mm I am sadly falling foul of the issues. My focus with the auto routines is always a fairly consistent amount away from good focus. I was thinking that in order to focus in Ha for example I have an offset - Is there a way to add an offset with the luminance filter that is actually focusing?

If this is doable, I could run the focus routine and add a +30 or so offset that I have found it always needs as a tweak to really nail that focus. Interestingly I have always found the focus to be out in the same direction, so this would definitely cure the problem.


Hi Sara,

Yes you can do this. I haven’t yet set this up, but I believe it’s done as follows:

You need to specify the offset for your Ha filter in the filters tab (control panel) so in your case it’s +30 [if I’m wrong with this bit I’m sure someone will correct me please !!]

Then in the focus tab (again, control panel) check ‘auto adjust focus per filter’. Finally, in the autofocus settings themselves, check ‘autofocus with filter’ and you can then select your Luminance filter.

Autofocus will then run with your luminance filter using whatever exposure time you set (much faster than trying to focus through the Ha filter itself). Your focuser should then move out say 30 steps from the focus point when your Ha filter is selected.

You can also save these settings to a profile given they’ll be different when imaging with a different scope…



Thanks for the reply - I’m sorry I didn’t make myself clear - I already have the offsets all sorted, what I want is the following.

I currently focus using my luminance filter and it is inaccurate. I want to know if there’s a way to focus as per usual in luminance and then add an offset to the luminance filter after it has been focused.


Ah, sorry, I’m with you now, but don’t know the answer to that one I’m afraid…


I guess this is not possible then? Anyone?


Hi Sara

I think you are using the 2.4 beta, is that correct? I have not upgraded to the beta and am still running the latest 2.3 release. In 2.3 the feature you mention above is not available. Perhaps it could be incorporated into 2.4 as a feature request? I’m sure either Ken or Jared will respond.

As for other suggestions, you may well already have experimented with my thoughts below, so apologies if I’m not adding anything new to your thinking and analysis:

  1. Have you tried excluding a percentage of the image frame, e.g. 5%, 10%, to see what effect that has on the achieved AF focus point vs grabber focus point? This is a feature in the auto focus set up (sorry, I don’t have SGP installed on my iMac at home, it’s only on my observatory pc, and can’t recall exactly where).
  2. The magnitude of the focus point error makes me think there might be a relation to backlash compensation. In another thread, (Intermittent Focus Fail . . . ) Ken said “This is true. Backlash is completely busted in 2.4 Beta. I would recommend turning it off right now.” Is it possible that your backlash compensation setting is simply not working? I have had problems with focus accuracy and repeatability when I’ve set this parameter incorrectly and also when the grub screw securing the stepper motor coupler to the focuser axle started working loose.
  3. Alternatively to (2), do you need to set a backlash compensation amount (although you can’t at the moment in the 2.4 beta release as it isn’t working)?
  4. Does the focus point with another filter, e.g. R, reach the point achieved by the grabber (or at least closer than the Lum filter)? If so, you could switch to this filter temporarily until the backlash compensation feature is fixed.

Good luck,



Another thought Sara, as a work around until you can figure out why there is a difference between the AF and grabber, if you are only using NB filters, if you have measured the offset for the Lum filter focus point vs grabber focus point, i.e. +30 steps, this figure could be added to the offset for each NB filter to create a ‘combined offset’. Thus, if the offset between the grabber Lum and grabber Ha is +25 steps, the offset from the Lum filter point then becomes +55. If the offset between grabber Lum and Grabber Ha is -25 steps, then the combined offset becomes +30-25, i.e. +5 steps.

Would that work do you think?



Thanks very much for your considered reply Barry. I was running the latest beta, but I had crashes and focus locks so that I couldn’t stop the AF running within the programme, so I went back to the older version!

I’ve managed to get it working to an acceptable standard at the moment, using settings that would frankly make peoples eyes water. You know that I am one to fly in the face of theory and rules and in doing so, I think I have reached a satisfactory point.

Regarding backlash - I’ve not got any set at all. The FT focuser on my scope is as tight as a drum and setting backlash just seemed to mess things up.

I like your idea about the NB offsets - This is something that I had thought about - I’ll see how it goes on another run.


Good stuff Sara. I just saw your post on SGL re SGP.

How frustrating for you though with your new scope. I like the Bok Globules you’ve posted, a contrast to the ‘softer’ images you were getting from the C9.25.



It’s unlikely that we will add separate software functionality that corrects another problem introduced by the same software (meaning we are averse to treating symptoms). That said, we are looking at several long FL / obstruction based AF options for 2.5. None of this will make it into 2.4 since that is now on feature lock in order to make it through the Beta process.



Just curious Sara but what settings did you use to get things to work?

I’m not 100% convinced about AF getting things absolutely spot on, but I’ve been running 30 min subs recently and something is messing up more of them up than I’d like. That said I’ve had the occasional night where 9 or 10 have all been bang on so I do think my woes are seeing and UK weather related as opposed to AF.

Whilst your RC set up is completely different to mine, I’d still be interested to hear of anything leftfield you may have tried !

Thanks .


The other night, because the AF routine is NOT getting me decent focus I spent some time working through a number of settings to reach the best focus. I had previously attained good focus with the Bahtinov mask and grabber, so know what figure I was aiming for. I then ran AF in different settings until I reached what was the closest.

I found that I got AF closest to decent focus with 4x4 bin, 70 steps and 8s exposures. I KNOW this shouldn’t work and isn’t the way that the manual says to do it, but this was based on testing with my setup.


Thanks for the reply - whatever works works, iirrespective of how its supposed to be !



I have tested SGP on 4 different focusers, ranging from low end GSO to mid end moonlite focuser, despite the fact that my moonlite focusers apparently don’t have backlash, setting a backlash value help to produce better results. With the GSO is a must to use backlash compensation.

It doesn’t hurt to have backlash compensation enabled, SGP implementation will work fine with focuser that don’t show backlash. Just set the backlash compensation 3 to 5 times your step size and try again.